# Ragas — Evaluate RAG & LLM Applications

> Ragas evaluates LLM applications with objective metrics, test data generation, and data-driven insights. 13.2K+ GitHub stars. RAG evaluation, auto test generation. Apache 2.0.

## Install

Save as a script file and run:

## Quick Use

```bash
# Install
pip install ragas

# Quick start — generate a RAG evaluation project
ragas quickstart rag_eval -o ./my-project

# Or evaluate in Python
python -c "
from ragas.metrics import faithfulness, answer_relevancy
from ragas import evaluate
# Evaluate your RAG pipeline outputs
# results = evaluate(dataset, metrics=[faithfulness, answer_relevancy])
print('Ragas ready! Define your dataset and run evaluate().')
"
```

---

## Intro

Ragas is a toolkit for supercharging LLM application evaluations with objective metrics, intelligent test data generation, and data-driven insights. With 13,200+ GitHub stars and Apache 2.0 license, Ragas provides LLM-based and traditional evaluation metrics for RAG pipelines (faithfulness, answer relevancy, context precision/recall), automatic test dataset generation from your documents, seamless integration with LangChain and observability tools, and feedback loops to leverage production data for improvement.

**Best for**: Teams building RAG pipelines who need systematic evaluation and quality measurement
**Works with**: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf
**Integrations**: LangChain, LlamaIndex, Langfuse, Phoenix

---

## Key Features

- **RAG metrics**: Faithfulness, answer relevancy, context precision/recall
- **Test data generation**: Auto-create evaluation datasets from documents
- **LLM + traditional metrics**: Both AI-judged and deterministic scoring
- **Production feedback loops**: Use real data to improve quality
- **LangChain integration**: Evaluate chains and retrievers directly
- **Async scoring**: Fast parallel evaluation with any LLM provider

---

### FAQ

**Q: What is Ragas?**
A: Ragas is an LLM evaluation toolkit with 13.2K+ stars. Objective metrics for RAG (faithfulness, relevancy), auto test generation, LangChain integration. Apache 2.0.

**Q: How do I install Ragas?**
A: `pip install ragas`. Quick start with `ragas quickstart rag_eval -o ./my-project`.

---

## Source & Thanks

> Created by [Exploding Gradients](https://github.com/explodinggradients). Apache 2.0.
> [explodinggradients/ragas](https://github.com/explodinggradients/ragas) — 13,200+ GitHub stars

---
Source: https://tokrepo.com/en/workflows/2c856b4d-64e5-46b2-9bbd-a7ce9f7a7296
Author: Script Depot